We will use a nomogram to identify a group of patients with clinically localized prostate cancer, at high risk for early disease progression. We have developed a nomogram for predicting the risk of failure (detected by rising serum PSA levels) following radical prostatectomy based on preoperative clinical characteristics (Gleason grade, clinical stage, serum PSA level, and the digital rectal examination of the prostate). According to this nomogram, the risk of recurrence is based on the total number of points that a patient accumulates. For example, patients with a score above 150 were in the upper quartile and had a 35% risk of prostate cancer recurrence (serum PSA failure) 5 years after radical prostatectomy. We will recruit our study participants from this group of patients with a high risk for recurrence (i.e., patients with a score of 150 and above). These patients must be in general good health with a life expectancy greater than 10 years who have chosen a radical prostatectomy for the treatment of their cancer. Following a discussion of the protocol, participants will be asked to sign an informed consent form. Patients will receive an ultrasound guided, intraprostatic injection of 1x1010 PFU of HSV-tk carrying adenoviral particles, through a 20 gauge needle. The injectable volume will be 1 ml. The exact technique of intra- prostatic injection will be reviewed after the first 4 patients and may need to be modified according to what we learn. All the patients will receive a broad spectrum oral antibiotic the evening before and the morning of the procedure, continuing the antibiotic twice daily for a total of 5 days. This injection will be performed in the SDU ultrasound laboratory. The patient will then be admitted to The Methodist Hospital for observation and insertion of a PIC intravenous line. Ganciclovir treatment will begin 24 hours post virus injection at 5 mg/kg I.V. over 1 hour every 12 hours for 2 weeks (28 doses). GCV treatment will be initiated in the hospital, but 24 hours later, patients suffering no ill effects will continue treatment at home and will be monitored by VNAs. The most commonly observed complication in patients receiving ganciclovir has been granulocytopenia and thrombocytopenia. These will be monitored on the second, seventh, and fourteenth day of treatment and as medically indicated. It is important to note that we did not observe these complications in the phase I study (maybe because our patients are in good health and immunocompetent. The literature data is based primarily on immunocompromised patients receiving this drug).